Bearing fault diagnosis based on SIR multistage residual connection dense network

Abstract

Abstract: In order to study the performance of extracting feature information from time-varying and weak signals of rolling bearings in rotating machinery under complex working conditions, a bearing fault diagnosis method based on SIR multi-stage residual connection dense network is proposed. Firstly, the SIR module is designed, which extracts more important and richer feature information by giving different weights to the input data feature channels and broadening the width of the network. Secondly, a multi-stage residual connection dense network is designed to adaptively extract the effective features from the bearing vibration signals. Finally, a softmax classifier is constructed to realize fault classification. Compared with other methods, the experimental results show that the proposed method can detect faults more accurately under variable noise, variable load, and variable working conditions, and has more robustness and generalization ability for complex working conditions.

Key words: fault diagnosis, rolling bearing, residual density network, feature recalibration, variable conditions

CLC Number:

TH133.33

ZHAO Xiao-qiang, LUO Wei-lan, LIANG Hao-peng. Bearing fault diagnosis based on SIR multistage residual connection dense network[J]. Journal of Lanzhou University of Technology, 2022, 48(6): 46-54.

References

[1] 鄢小安.基于数学形态学的滚动轴承故障诊断方法研究 [D].南京:东南大学,2019.
[2] WANG Xiang,ZHENG Yuan,ZHAO Zhenzhou,et al.Bearing fault diagnosis based on statistical locally linear embedding [J].Sensors,2015,15(7):16225-16247.
[3] CAESARENDRA W,WIDODO A,YANG B-S.Application of relevance vector machine and logistic regression for machine degradation assessment [J].Mechanical Systems and Signal Processing,2009,24(4):1161-1171.
[4] SAIDI L,FNAIECH F,HENAO H,et al.Diagnosis of broken-bars fault in induction machines using higher order spectral analysis [J].Isa Transactions,2013,52(1):140-148
[5] 胡爱军,赵军,孙尚飞,等.基于谱峭度和最大相关峭度解卷积的滚动轴承复合故障特征分离方法 [J].振动与冲击,2019,38(4):158-165.
[6] 程发斌,汤宝平,刘文艺.一种抑制维格纳分布交叉项的方法及在故障诊断中应用 [J].中国机械工程,2008(14):1727-1731.
[7] 祝小彦,王永杰,张钰淇,等.基于自适应最优Morlet小波的滚动轴承故障诊断 [J].振动.测试与诊断,2018,38(5):1021-1029,1085.
[8] PICHLER K,LUGHOFER E,PICHLER M,et al.Fault detection in reciprocating compressor valves under varying load conditions [J].Mechanical Systems and Signal Processing,2016,70/71:104-119.
[9] WANG Liming,SHAO Yimin.Fault feature extraction of rotating machinery using a reweighted complete ensemble empirical mode decomposition with adaptive noise and demodulation analysis [J].Mechanical Systems and Signal Processing,2020,138: 106545.
[10] JIANG Liying,LI Qianqian,CUI Jianguo,et al.Rolling bearing fault diagnosis based on Higher-order cumulants and BP neural network [C]//27th Chinese Control and Decision Conference.[S.l.]:IEEE,2015:2683-2686.
[11] 刘畅,伍星,刘韬,等.基于近似等距投影和支持向量机的滚动轴承故障诊断 [J].振动与冲击,2018,37(5):234-239.
[12] MURALIDHARAN V,SUGUMARAN V.A comparative study of Nave Bayes classifier and Bayes net classifier for fault diagnosis of monoblock centrifugal pump using wavelet analysis [J].Applied Soft Computing,2012,12(8):2023-2029.
[13] WANG Yalin,YANG Haibing, YUAN Xiaofeng ,et al.Deep learning for fault-relevant feature extraction and fault classification with stacked supervised auto-encoder [J].Journal of Process Control,2020,92:79-89.
[14] WANG Hongmei,LIU Pengzhong.Image recognition based on improved convolutional deep belief network model [J].Multimedia Tools and Applications,2020,80(9):1-15.
[15] 周奇才,沈鹤鸿,赵炯,等.基于改进堆叠式循环神经网络的轴承故障诊断 [J].同济大学学报(自然科学版),2019,47(10):1500-1507.
[16] 车畅畅,王华伟,倪晓梅,等.基于深度学习的航空发动机故障融合诊断 [J].北京航空航天大学学报,2018,44(3):621-628.
[17] 温江涛,闫常弘,孙洁娣,等.基于压缩采集与深度学习的轴承故障诊断方法 [J].仪器仪表学报,2018,39(1):171-179.
[18] 陶洁,刘义伦,付卓,等.基于Teager能量算子和深度置信网络的滚动轴承故障诊断 [J].中南大学学报(自然科学版),2017,48(1):61-68.
[19] 张弘斌,袁奇,赵柄锡,等.采用多通道样本和深度卷积神经网络的轴承故障诊断方法 [J].西安交通大学学报,2020,54(8):58-66.
[20] 张西宁,刘书语,余迪,等.改进深度卷积神经网络及其在变工况滚动轴承故障诊断中的应用 [J].西安交通大学学报,2021,55(6):1-8.
[21] 陈保家,陈学力,沈保明,等.CNN-LSTM深度神经网络在滚动轴承故障诊断中的应用 [J].西安交通大学学报,2021,55(6):28-36.
[22] HU J,SHEN L,ALBANIE S,et al.Squeeze-and-excitation networks [J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2020,42(8):2011-2023.
[23] SZEGEDY C,LOFFE S,VANHOUCKE V,et al.Inception-v4,inception-resnet and the impact of residual connections on learning [C]//Thirty-First Conference on Artificial Intelligence.San Francisco:[s.n.],2017:4278-4284.
[24] ZHANG Zhicheng,LIANG Xiaokun,DONG Xu ,et al.A sparse-view CT reconstruction method based on combination of densenet and deconvolution [J].IEEE Transactions on Medical Imaging, 2018,37(6):1407-1417.
[25] ZHANG Yulun,TIAN Yapeng,KONG Yu,et al.Residual dense network for image super-resolution [C]//2018 IEEE/CVF Conference on Computer Vision And Pattern Recognition.[S.l.]:IEEE,2018:2472-2481.
[26] 孙俊,何小飞,谭文军,等.空洞卷积结合全局池化的卷积神经网络识别作物幼苗与杂草 [J].农业工程学报,2018,34(11):159-165.
[27] SMITH W A and RANDALL R B.Rolling element bearing diagnostics using the Case Western Reserve University data:A benchmark study [J].Mechanical Systems and Signal Processing,2015,64/65:100-131.
[28] 赵小强,梁浩鹏.使用改进残差神经网络的滚动轴承变工况故障诊断方法 [J].西安交通大学学报,2020,54(9):23-31.